Hi Researchers and Scholars,
I recently just managed to compute my survey and came out with a result via SmartPLS. All my constructs are reflective*
I have a few questions to ask regarding my result.
1) I have two latent variables which have loadings less than 0.700 (two out of three to be exact). I have read several forums and journals stating that if i remove one loading >0.700 and the AVE and composite reliability is more than 0.5 or 0.7 respectively i can keep the 2nd loading (which is less than 0.700). Is it right to say so?
2) I have one path loading (give it a name A to B) which was suppose to be negative in my literature review but in SmartPLS algorithm, it turns out that the path coefficient was positive and the t-value was significant. Other than reverse coding it what else can i do to resolve this issue?
It will be of great help for anyone who can assist me with this issue. Thank you so much.
Interpreting my result in SmartPLS. Please Help
-
- PLS Junior User
- Posts: 2
- Joined: Fri May 26, 2017 11:11 am
- Real name and title: Benjamin Chua
- cringle
- SmartPLS Developer
- Posts: 818
- Joined: Tue Sep 20, 2005 9:13 am
- Real name and title: Prof. Dr. Christian M. Ringle
- Location: Hamburg (Germany)
- Contact:
Re: Interpreting my result in SmartPLS. Please Help
@1: Only delete items with outer loadings below 0.4. If loadings are below 0.7, only delete them if necessary to increase the AVE above 0.4 and the composite reliability above 0.7.
@2: the sign depends on the orientation of your items and your construct. re-scaling could be an option.
Best
CR
@2: the sign depends on the orientation of your items and your construct. re-scaling could be an option.
Best
CR
Prof. Dr. Christian M. Ringle, Hamburg University of Technology (TUHH), SmartPLS
- Literature on PLS-SEM: https://www.smartpls.com/documentation
- Google Scholar: https://scholar.google.de/citations?use ... AAAJ&hl=de
- Literature on PLS-SEM: https://www.smartpls.com/documentation
- Google Scholar: https://scholar.google.de/citations?use ... AAAJ&hl=de